# Osmosis

## Docs

- [Best Practices](https://docs.osmosis.ai/git-sync/best-practices.md): Best practices and troubleshooting for Git Sync
- [Folder Structure](https://docs.osmosis.ai/git-sync/folder-structure.md): Required repository structure for Osmosis Git Sync
- [MCP Tools](https://docs.osmosis.ai/git-sync/mcp-tools.md): Creating MCP tools for Osmosis
- [Set Your Own Repo](https://docs.osmosis.ai/git-sync/overview.md): Sync your GitHub repository with Osmosis AI platform
- [Reward Functions](https://docs.osmosis.ai/git-sync/reward-functions.md): Creating deterministic reward functions for LLM evaluation
- [Reward Rubrics](https://docs.osmosis.ai/git-sync/reward-rubrics.md): Using LLMs to evaluate outputs with natural language criteria
- [Introduction](https://docs.osmosis.ai/introduction.md): Osmosis is a comprehensive post-training platform that helps companies create task-specific models beating foundation models at a fraction of the cost.
- [GitHub Integration](https://docs.osmosis.ai/platform/github-integration.md): Connect your GitHub repository for automatic sync of tools, rewards, and rubrics
- [Monitoring & Metrics](https://docs.osmosis.ai/platform/monitoring.md): Track training progress with real-time metrics, logs, and checkpoint management
- [Platform Overview](https://docs.osmosis.ai/platform/overview.md): What is Osmosis Platform and what can you do with it
- [Getting Started](https://docs.osmosis.ai/platform/quickstart.md): Sign up, create a project, and start training
- [Training Runs](https://docs.osmosis.ai/platform/training-runs.md): Create, configure, and manage reinforcement learning training runs
- [Workspace & Team](https://docs.osmosis.ai/platform/workspace-settings.md): Manage your workspace, team members, API keys, and integrations
- [API Reference](https://docs.osmosis.ai/python-sdk/api-reference.md): Complete Python SDK API reference for Osmosis AI
- [CLI Reference](https://docs.osmosis.ai/python-sdk/cli-reference.md): Complete reference for the osmosis-ai command-line interface
- [Python SDK](https://docs.osmosis.ai/python-sdk/installation.md): Install and configure the Osmosis AI Python SDK
- [Agent Loop Guide](https://docs.osmosis.ai/remote-rollout/agent-loop.md): Implement custom agent loops for Osmosis training
- [Eval Mode](https://docs.osmosis.ai/remote-rollout/eval-mode.md): Evaluate trained models with custom eval functions and pass@k metrics
- [Remote Rollout Overview](https://docs.osmosis.ai/remote-rollout/overview.md): Build custom agent loops for Osmosis training infrastructure
- [Quick Start](https://docs.osmosis.ai/remote-rollout/quickstart.md): Build and test a remote rollout agent in 5 minutes
- [Test Mode](https://docs.osmosis.ai/remote-rollout/test-mode.md): Test agent loops locally with external LLM providers

## OpenAPI Specs

- [openapi](https://docs.osmosis.ai/api-reference/openapi.json)